Geoscience Reference
In-Depth Information
containing soft sediments and related local lithological and topographic conditions.
However, geology is not the only factor which explains the observed elongation of
the intensity patterns. Directivity effects played an important role in the damage am-
plification particularly for intensities 8, 9 and 10. This is attested by the PGA records
which show higher values for the E-W components than the N-S ones, independently
from station azimuth, epicentral distance and site conditions (Laouami et al. 2006).
As the first step of analysis, we compared the damage distribution with the pop-
ulation density inside the provinces where the event was felt (Appendix 1). The
2003 epicentral area that includes a large section of the eastern suburb of Algiers
city and related large population within the Mitidja basin is crossed by at least 3
macroseismic curves. The macroseismic field investigations also indicate that dam-
age distribution is strongly conditioned by soil conditions and building vulnerability.
In some zones, the separation of the isoseismals X and IX could not be achieved be-
cause of the large variation of the local geology and likely related site effects. The
isoseismals are clearly asymmetric and elongated in the NE-SW direction which
represents the fault rupture strike.
5 Attenuation
The assessment of seismic hazard at a given site requires an attenuation law for
the peak ground acceleration (PGA). The intensity-attenuation relationship is ob-
tained by deriving empirical correlations between intensity and epicentral distance
for earthquakes for which isoseismal maps are available. Many authors developed
attenuation relationships worldwide. Douglas (2001) presents a valuable summary
of 121 published attenuation relations for PGA. Examples include the attenuation
laws developed by Idriss (1978), Joyner and Boore (1981), Campbell (1985), Boore
and Joyner (1982), Joyner and Boore (1988), Sadigh et al. (1993), Ambraseys and
Boomer (1991) and Ambraseys (1995). In Algeria, generally when assessing the
seismic hazard at a given site, authors (Benouar 1996, Naili and Benouar 2000,
Laouami et al. 2004) adopt the PGA attenuation laws developed by Ambraseys and
Boomer (1991) (Equation (1)) and Ambraseys (1995) (Equation (2)). The former has
been derived from 529 accelerograms recorded mainly on soft rock or soil from 219
shallow seismic events (
25 km) mainly in the Mediterranean region, which includes
Algerian data, and the second is based after data correction on 1,260 accelerograms
generated from 619 shallow seismic events of which 3% are Algerian data.
≤
log
10
(
a
h
)
=−
0
.
87
+
0
.
217(
Ms
)
−
log
10
(
r
)
−
0
.
00117(
r
)
±
0
.
26
P
(1)
log
10
(
a
h
)
=−
1
.
43
+
0
.
245(
Ms
)
−
0
.
786 log
10
(
r
)
−
0
.
0010(
r
)
±
0
.
24
P
(2)
Where r
2
(d
2
h
2
), h is the focal depth (taken at an optimum value of h
7km),
d is the epicentral distance in km, Ms is the surface-wave magnitude, and a
h
is
the predicted peak horizontal ground acceleration. The values 0.26 and 0.24 in
Equations (1) and (2) are the respective standard deviation. The parameter P takes a
=
+
=
2
.
Search WWH ::
Custom Search